Intellectual Territories in Anglo-Saxon England

This is a repository copy of What were they thinking? : Intellectual territories in Anglo- Saxon England.

White Rose Research Online URL for this paper: https://eprints.whiterose.ac.uk/102096/

Version: Accepted Version

Book Section: Carver, Martin orcid.org/0000-0002-7981-5741 (2011) What were they thinking? : Intellectual territories in Anglo-Saxon England. In: Hinton, David A, Crawford, Sally and Hamerow, Helena, (eds.) The Oxford Handbook of Anglo-Saxon Archaeology. Oxford University Press , Oxford , pp. 914-947.

Reuse Items deposited in White Rose Research Online are protected by copyright, with all rights reserved unless indicated otherwise. They may be downloaded and/or printed for private study, or other acts as permitted by national copyright laws. The publisher or other rights holders may allow further reproduction and re-use of the full text version. This is indicated by the licence information on the White Rose Research Online record for the item.

Takedown If you consider content in White Rose Research Online to be in breach of UK law, please notify us by emailing [email protected] including the URL of the record and the reason for the withdrawal request.

[email protected] https://eprints.whiterose.ac.uk/ Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:56 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

chapter 47 ...... WHAT WERE THEY THINKING? INTELLECTUAL TERRITORIES IN ANGLO-SAXON ENGLAND ......

martin carver

INTRODUCTION ......

Over the past 150 years the study of the material culture of south-east Britain between ad 400 and 1100—the study we call Anglo-Saxon archaeology—has largely focused on the illustration of a historical model based on written documents, inscriptions, and place-names. This model sees the arrival of unspecified numbers of Germanic incomers during the fifth century into an area occupied by partially Christianized Britons and Roman citizens of diverse origin. During the fifth to seventh centuries, the incomers became culturally and linguistically dominant and created new kingdoms (Kent, East Anglia, Northumbria, Wessex) within broad tribal affiliations (Angles, Saxons,and Jutes). The creation of the kingdoms was coincident with political pressure from the Continent and the arrival of a new Christian mission from the south, although it is not clear which of these initiatives was primary or determinant. The instruments of the Christian kingdoms were kings, operating from palaces (villae regales) and stimulating trade through market Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:56 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 933

centres (wics), and the ruling families were closely allied to a Christian network based on monasteries (monasteria). Through the seventh and eighth centuries this confederacy built up alliances and exchanges with other Christian kingdoms on the Continent and in Mediterranean lands. But in the early ninth century it suffered widespread disruption when Scandinavian marauders targeted monasteries either out of ideological revulsion or economic exclusion, or both. Their initial reign of terror subsided into settlement and the governance of the country was divided between the Danelaw (north) and the kingdom of Wessex (south). In the late ninth /early tenth century Alfred of Wessex and his successors Edward and Æthelflæd set about the conquest or reconquest of the greater part of Britain, using a new type of fortified town (burh). This resulted in the creation of a kingdom of England which was broadly located within its modern boundaries. During the tenth century the church was institutionalized so as to mobilize the full framework of diocese, parish, reformed monastery, and secular college. However, ambition to control the wealthy island of Britain and collect its revenues was prosecuted by the south Scandinavians, the Normans, and later the French well into the Middle Ages (Chadwick 1905; Levison 1946; Whitelock 1954; Bassett 1989; Higham 1995; Fletcher 1997; Blair 2005). Anglo-Saxon archaeology has been highly successful in illustrating this narrative. The large-scale excavation of more than a hundred cemeteries has shown that burial assemblages and rituals noted in the south-east parts of Britain correlate well with Germanic practice in northern mainland Europe and with pagan ideas that survived in later medieval Scandinavian literature. Archaeologists have excavated convincing examples of palace sites, wics, and monasteries belonging to the mid Saxon period, and burhs of the later Saxon period, enlarging and refining their economic and symbolic roles through studies of finds assemblages and the use of space. The Christian liturgy and its development are recognizable in church buildings, sculpture, and illuminated manuscripts, all of which also contribute to an enhanced picture of the metaphysics and artistry of the Anglo-Saxon people (Hodges 1989; Carver 1992; Welch 1992; this volume, passim). With such a record of achievement, it would be understandable if scholars of an interdisciplinary persuasion were content to continue to celebrate and research the Anglo-Saxons through the discussion of objects, buildings, and sites which are already linked to their world through Latin or Old English literature. But, as this handbook shows, the twenty-first century presents us with a new agenda and new principles for Anglo-Saxon archaeology. These can be attributed to three different forces, the first of which is the rise of theoretical archaeology and its increasing influence on the early medieval period, something not unconnected with its administrative academic grouping with prehistory rather than history departments (Andreń 1998; Carver 2002b; Hills 2007). Archaeologists applied the study of process to the emergence of the Anglo-Saxon kingdoms (e.g. Carver 1989; Scull, this volume) and structuralism to the complexity of metalwork styles in the early Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:56 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

934 m. o. h. carver

period (Gaimster, this volume). Post-structuralist thinking invited us to consider the effects of the current political climate on our own research, and this has led, directly or indirectly, to the questioning of assumptions about ethnicity, gender, and religion—particularly the degree to which they are signalled in graves (for histories of thinking see Rahtz et al.. 1980; Lucy 2000: ch. 1; and Williams 2005). As Heinrich Ha¨rke (2007) has pointed out, this is not an evolving topic, but a cyclical one. In an article published posthumously in 2005, Bartholomew expressed the conviction that through the archaeological tracking of migrations (which he sees as coming from Frisia) ‘it is possible to recover, once again, a true understand- ing of the mighty movement of peoples which took place in the early fifth century and which transformed the Britannia of the Late Roman empire into the land of the Angles, the Engla-land in which we live today’ (2005: 28). Ha¨rke (2002: 147) puts it bluntly: ‘The few historical sources depict the process of settlement in terms of what would now be called “ethnic cleansing” of the native Britons by large-scale immigration of Anglo-Saxons.’ Alex Woolf, writing around the same time, comments: ‘Rejection of the obviously legendary accounts of the fifth-century invasions dates back to Kemble, but more recently it seems as if some scholars believe that proving that the invasions did not happen as they were described in later sources is the same thing as proving that they did not happen at all’ (2000: 99). Myres in 1936 felt that these were times ‘whose quality cannot be portrayed without serious distortion in those broad and rational sequences of cause and effect so beloved by the historian. The conflicts are too complex, the issues too obscure, the cross-currents too numerous, and the decisions too local, to make possible the application of any single formula to their solution; and it is at least reassuring to remember that, if we found such a formula, we should unquestionably be wrong’ (Myres in Collingwood and Myres 1936: 455–6). The earliest of these authorities is thus perhaps closest to the scepticism of today. In the critique of their historical model, recent Anglo-Saxon archaeologists have shown some aversion to the acceptance of overwhelming Germanic immigration and pagan thinking, preferring to promote the influence of Britons or British culture (Lucy 2000; Hills 2003). The evidence for British continuity on the ground is still elusive and indirect—the continuation of farming on the same piece of land, the proximity of Roman settlements and Anglo-Saxon cemeteries, or the presence of stones and planks in an unfurnished grave (Rahtz et al. 2000). Moreover the initial scholarly reservations have been inflated in cavalier hands into a generalized negativity in which there were no invasions and Britain was always continuously occupied by Britons throughout pre- and proto-history (e.g. Pryor 2004: 176). Moreland (2000: 50) comments ‘Given the lack of evidence for a massive rupture in the countryside of Late Roman Britain, however, we must assume that the vast majority of the population was not of Germanic origin/descent.’ The irony of such assertions is that one ethnic assumption is simply replaced by another. Meanwhile Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:56 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 935

the massive culture shifts that took place in the fourth to seventh centuries remain unexplained. A second key influence is provided by new tools in the scientific toolbox which to some extent have allowed archaeologists to break free of traditional documentary and artistic forms of dating and provenance, with their unfortunate propensity for circular argument. Radiocarbon dating, assisted by Bayesian analysis, offers a date of death for a buried person, whether inhumed or cremated, to within fifty years (Buck et al. 1996). Correspondence analysis brings greater precision to the typological progression of both objects and assemblages (Hines et al. 1999; Brug- mann and Penn 2007). Stable isotope analysis has begun to report on the likely birthplaces of buried persons, so contributing to the migration debate, if only anecdotally. At West Heslerton, for example, four out of twenty-four individuals had oxygen values suggesting a childhood in ‘eastern continental Europe or, more likely, Scandinavia’. These were also the only females to be buried without brooches, in poorly furnished graves. Of the twenty others (fourteen females, five males) seven were local and thirteen originated west of the Pennines (Budd et al. 2004; though see Hull and O’Connell, this volume). These results raise fresh, and perhaps more interesting, questions than those of migration. Who were these westerners, and what made them seek (or fail to escape) a life in Yorkshire? Population mobility, while it certainly involves settlers from overseas, must also reflect marriage, slavery, conscription, the hire of agricultural labour, and every other form of human trafficking. Stable isotope analysis is more likely to give us a hundred small stories than one big one—although none the worse for that. DNA analysis, once it is reliably applicable to ancient material, will also help to group ancient peoples; it is currently a blunt instrument using the modern population as its point of departure (Thomas et al.. 2006; Hills 2003). The third element of the new agenda is more positive in outlook and is the subject of the rest of this chapter. It combines three aspects of recent theory as its foundation. In Braudel’s layered history, the ecological and environmental circumstances (the longue dureé), social circumstances, and personal circumstances all change at different rates (Braudel 1972). Peer Polity Interaction (PPI) assumes that social groups have an equal chance of influencing each other, as compared to core and periphery in which one group is dominant (Renfrew and Cherry 1986). Thus PPI may be thought particularly appropriate to the period between the Roman Empire and the Middle Ages, in which, during a period almost unique in European history, no overarching authority succeeded in holding sway. These two frame- works of ideas have been found very helpful in the interpretation of material culture. The third, a relative newcomer to the party, takes as its underlying basis the assumption that material culture can be ranked in terms of its investment, such that the greater the investment, the greater will be the intended expression. In any community, objects of relatively high investment—defined as ‘monuments’—have messages that relate to the community’s view of itself at a particular moment. Thus Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:56 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

936 m. o. h. carver

monuments do not report on a specific social structure, ethnicity, or religion, which indeed they are hardly in a position to know; but they do offer an epitome of a contemporary, short-lived, and local thinking. It may well be possible one day to collate examples of these encapsulated thoughts into a narrative history, just as it would be possible in theory to write a general history from a thousand pages torn from individual diaries. But for the present the archaeological inquiry mainly chronicles the results of encounters with a wide range of individuals and records their opinions and aspirations (Carver 2001; 2003; Frazer in Frazer and Tyrrell 2000: 4). This kind of study requires a different kind of analytical language to the one that gave us the historical model with which this chapter began. We now accept that we cannot declare a grave to be Christian or Saxon, but we can recognize that it makes references or allusions to cultural material that we already know. Unfortunately, the material concerned has already been ethnically pre-labelled during decades of study, so we may be obliged to use terms such as ‘culturally Saxon’, ‘culturally female’,or ‘culturally Christian’ to show we are talking about a type of object, a type of burial, or a way of thinking, but not about a type of person. The mixed cultural messages that are features of early medieval monuments and led to the questioning of ethnic and religious attributions in the first place are here seen as a strength. The underpinning assumption is that the messages are not mixed because of confusion or unfamiliarity, but because early medieval people are dextrously employing a common vocabulary to say something creative and original. The aim of the archaeologist is to record and interpret myriad local voices, without expecting them to bear directly onto a pre-existing historical framework—although they may. In our analysis of sites and monuments we find we cannot often reliably distinguish migration, trade, diplomacy, proselytization, sex, rank, or documented forms of religion or ethnic groups. But we can use the detected references and allusions to define identities and ideologies, which signify allegiance or defiance to contemporary forces. Given that we are dealing with the construction of objects or graves or buildings or sites, what we observe is the active signalling of adherence to a world view and a future programme, in other words, politics. In this line of reasoning, what the larger monuments mainly signify is political alignment. To show this kind of approach in action, or rather to illustrate its potential, there follow four case studies of sites, one from each early medieval sub-period, which attempt to break down monuments into their political alignments. A second group of case studies considers how artefacts may be used to define territories, and what these territories might mean. Some attempt will then be made to compare and contrast these archaeological alignments and territories with those known from documents. It will be found that while archaeology cannot always endorse the historical record in the way we might wish, it can open its own window onto the Anglo-Saxon experience. Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:56 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 937

WASPERTON: AN INTELLECTUAL FRONTIER OF THE FOURTH TO SEVENTH CENTURIES ......

The cemetery at Wasperton was in use from the fourth to seventh centuries and lay within a prehistoric landscape, ten hectares of which were excavated (Fig. 47.1). A programme of radiocarbon dating and stable isotope analysis indicated likely dates of death and regions of birth for a handful of examples, and the combination of radiocarbon dating, stratification, and grave alignment allowed the construction of a sequence in which all graves, furnished and unfurnished, could be included. The prehistoric landscape in the immediate vicinity included a Neolithic earthwork, a Bronze Age round barrow, and Bronze Age, Iron Age, and Roman enclosures and field boundaries, many of which were visible to the users of the cemetery. The first burials are those of members of a third- to fourth-century agri-business who commandeered a Romano-British enclosure for the purpose. These burials were well spread out so as to imply family plots, and diagnostic burial rites of the period included decapitation, north-south orientation, and inhumations dressed with neck-rings, bracelets, and hobnails. Of five fourth-century individuals, three grew up locally and two can be assigned a Mediterranean childhood. Throughout most of the fifth century, graves were unfurnished and aligned north- south or west-east; a significant number included stones and planks, and there was a notable plot in the south-east quadrant of the enclosure in which this type of burial predominated. Of nine burials assigned to the fifth century and analysed for stable isotopes, five grew up locally, three were from the West Country, and one was raised in a Mediterranean land. During the fifth century a group of cremations made its appearance. One urn was buried in a Neolithic earthwork, but the majority was clustered in the western half of the enclosure and demarcated by an arc of post- holes. Their cremation urns were paralleled in East Anglia and in north Germany. The first furnished inhumations of Germanic type appeared at much the same time, while during the sixth century, furnished, gendered inhumation—equipped with weapons or lavish ornamental dress items—was the diagnostic and dominant burial rite. In the earlier sixth century the artefacts appeared to reference East Anglia and the upper Thames valley; but in the later sixth century, the references were only southerly, to the Thames Valley as a whole and to the territory of Wessex beyond (Fig. 47.2). In the mid sixth century, burial for certain individuals, male and female, moved outside the enclosure on its northern side, where they were placed around or within one pre-existing and one newly-built mound. Of five burials assigned to the early seventh century, four were relatively well-furnished and placed at the north, east, west, and south sides of the now almost erased enclosure. A fifth grave, dated to the early seventh century by radiocarbon, was lined with planks and stones and placed in the south-east quadrant of the enclosure. This individual had a Mediterranean provenance (Carver et al. 2009). Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:56 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

938 m. o. h. carver

4th century 5th century

5-6th century 6th century

later 6th century 6-7th century

0 50m Figure 47.1 Cemetery evolution at Wasperton, fourth to seventh centuries (Carver et al. 2009: ﬁg. 1.2) Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:57 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

Applied brooch with six masks Equal armed brooch Applied brooch with star face Buckle with kidney shaped loop (Type I.8) Cast saucer brooch with four spirals Small symbol = 1 example Square-headed brooch of Hines early Gp 2 Large symbol = 2–5 examples Small-long brooch with lozenge foot

Shield boss with disc applique Buckle with rectangular plate (type II.24b) Square-headed brooch of Hines late Gp 2 Saucer brooch with running leg Saucer brooch with seven spirals Saucer brooch with five-point star Button brooch Small symbol = 1 example Large symbol = 2–5 examples Saucer brooch with quatrefoil

Figure 47.2 Changes of alignment: Distribution of artefacts that occur at Was- perton (a) in the earlier sixth century and (b) the later sixth century (Carver et al. 2009: ﬁgs. 4.22–3) Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:57 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

940 m. o. h. carver

This sequence can be read without recourse to immigration, ethnicity, or Chris- tianity, without denying that such attributes may one day be recognizable. For the time being we can note that during the fourth century a cemetery was established in family plots to serve an ideologically mixed population. Some aspects of burial rites (the hobnails and decapitation) reflect the wider Roman world, but others, such as north-south and west-east orientation, are following a local preference. Whereas we are not entitled to assume that all, some, or none of these burials reflect a Christian community, we may certainly claim the presence of a variety of ideologies that we do not otherwise know. This is also true of the west-east fifth-century burials with planks and stones, about which we can note that they include westerners and reflect not only practice in contemporary western Britain but a fainter echo of previous prehistoric burial. By contrast, the type and location of the fifth-century cremations strongly indicate an intrusive group, and the allusions they make are to East Anglia or to northern continental Europe. The inhumations were acculturated into the former cemetery plots, presumably as a result of marriage, but the accentuation of gender and the signals of alignment with the northern continent grow stronger. Barrow burial from the mid sixth century may or may not imply an increase in ranking, but it does imply a new fealty to the still visible prehistoric landscape. This is unlikely to imply a ‘past regained’ or a legitimation of land (already firmly possessed), but rather an ideological swing towards whatever meaning barrows were then thought to have (cf. Bradley 1987;Carver2002a; Mu¨ller-Wille 1992, 1993). We can also note that the development of the cemetery always included unfurnished graves in every plot, but the furnished burial rite barely intruded into one conservative corner, where the west- east plank burials continued from the Roman period to the seventh century. On the face of it, this is a Roman, British, and Anglo-Saxon cemetery and it displays versions of Christianity and other religions running alongside each other, as expressed in burial practice. The archaeology is really reporting the way people were thinking, rather than the way they were acting. Thus, the political alignment turns this way and that, but even in this small group of fewer than fifty persons alive at the same time, they never spoke with a single voice. This cemetery, like others, is not a fuzzy expression of migration, or of acculturation between pre-defined religions or ethnic groups, but a rather precise expression of complex ideological identities not yet in the history books, ideas that we have only just begun to study.

SUTTON HOO: CHANGES IN ALIGNMENT DURING THE SIXTH TO TENTH CENTURIES ......

The earliest Anglo-Saxon burials deﬁned at Sutton Hoo belong to a sixth- to seventh-century cemetery excavated in the year 2000 in advance of the construction of a visitor centre (Carver 2005: ch. 13). The latest burial rites include small Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:57 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 941

earth mounds and a cremation in a bronze bowl, a harbinger of the earliest rituals at the better known aristocratic burial ground which lies 300 m to the south (Fig. 47.3). Here, traces of prehistoric earthworks (actually field systems) were visible on a promontory overlooking the River Deben, where up to eighteen mounds were subsequently erected, the earliest at the corners of the Iron Age earthworks. In the hectare excavated, the burials began with the cremations of men, originally well furnished, accompanied on the pyre by sheep, horse, and cattle in around ad 590 (Fig. 47.4; Mounds 5, 6, 7). Next, a young man was buried in a coffin, with his horse in a separate pit beneath the same mound (Mound 17). Then, in the 620s came two ship burials, the first in which the ship was placed over the top of a well-furnished chamber (Mound 2) and the second where the chamber was constructed inside a ship placed within a trench underground. The horse-and-rider burial and the two ship burials were all wealthy weapon graves, provisioned also with feasting equipment. The Mound 1 chamber also famously contained a helmet, shield, standard, and sceptre all richly ornamented with symbolic images. This was the apogee of the princely burial ground, but there was sporadic late use in the form of modest graves for three young persons and the chamber grave of a high- status woman bedecked with silver (Mound 14). The princely burial ground only endured for some sixty years (c.590–650). It was, however, re-used almost immediately for the burial of execution victims who had been hanged on a gallows or displayed on a gibbet at one site to the east of the mounds and another on top of Mound 5. The princely burial ground was used as a place of execution, at a time which radiocarbon dates place between the eighth and the tenth centuries, by which time a Christian governance of East Anglia is not in doubt. Whether it represents the dispatch of villains, dissidents, sinners, or non- conformists, this ritualized killing field was the work of Christian kings (Carver 1998: 2005: 315–59). The Sutton Hoo burial ground has been reconciled with the exiguous written records to tell a story of kingship and conversion—and it may be so. However, the type of archaeology considered in this chapter focuses on the history of the intellect. During its short life, the princely burial ground was the scene of investment of outstanding intensity. The cremations in bronze bowls refer not only to the family cemetery a few hundred yards downriver, but to earlier burial practice in north Germany. Horse-and-rider graves begin in the Rhineland where they have been suggested to originate with a Frankish warrior group (Mu¨ller-Wille 1970–1; 1998). Ship-burials of this and earlier periods are known from the Baltic and the continental North Sea coast (Mu¨ller-Wille 1974, 1995; Carver 1995). In England all these burial rites, which refer to an earlier period overseas, are confined to the same part of the country, namely the Sandlings of Suffolk, which also features most of the known lyre burials (Lawson in Filmer-Sankey and Pestell 2001: 215–23). What- ever this flowering of special celebration might mean, we can at least be certain that it was in no sense traditional, or a ‘final phase’ (see Geake 1997). These are new Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:57 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

942 m. o. h. carver

[MELTON]

[WILFORD BRIDGE] (Site of ford)

Gallows (12th century)

[WOODBRIDGE] Tranmer House Bromeswell bucket cemetery

“Haven”

Possible burial mound

RIVER Gallows DEBEN (8th century)

SUTTON HOO BURIAL MOUNDS 20m AOD Contour

30m AOD Contour 10m AOD Contour

Burials excavated in Tranmer House cemetery Geophysical anomalies (possible burials) 0 m 500 m Possible portage routes

Figure 47.3 Location of Sutton Hoo by the River Deben in Suffolk. Tranmer House cemetery marks the site of Sutton Hoo’s sixth- to seventh-century predecessor. (Carver 2005: ﬁg. 220) Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:57 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 943

INT 41 2 0 50 m 12

17 14 5 20 18 INT 44 INT 48 INT 50 6 INT 32 Group A 7

1 13 INT 55

10 3 8

15 9 11 Group B

Inhumation in burial chamber in ship Inhumation under mound Inhumation in burial chamber under ship Cremation in bronze bowl Child‘s grave Cremation on oak tray dugout Execution burials Cremation Figure 47.4 Plan of the Sutton Hoo cemetery, sixth to tenth centuries (Carver 2005: ﬁg 219)

burial rites with a new message that obtained only in a speciﬁc part of England at a particular time. It may of course also refer to other processes such as kingship, or apprehension at the approach of the Christian missions, but we are not offered direct access to such historical matters, any more than there is information on whether any documented ﬁgures lie buried there (Carver 2001; Bruce-Mitford 1975: ch. 10; Carver 2005: 502–3 for critique). Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:57 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

944 m. o. h. carver

In this reading, the Mound 1 ship burial is neither the last of a great tradition nor a template for the burial of Anglo-Saxon kings. It is a ‘palimpsest of allusions’, allusions to the Roman empire, to Scandinavian heroes, to the island of Britain, to the Northern gods, to the Celtic gods, to Christ, and so forth. In this it resembles a poem in which the hopes and fears of the day were reiﬁed as appropriate to one place, person, and occasion (Carver 2000; cf. Frank 1992). In this sense the properties of the burial are not even cultural; they are original. It must be admitted that to produce a convincing reading of such a poem, as with Beowulf, will take much more study. The link with history is not hopelessly unreadable, but it must be reﬁned. Looking at ship burial, for example, we cannot trace it as an adoptive practice relayed from one place to another. It comes from a shared intellectual inheritance, and while we may not be in a position to decide what ship burial means, we can perhaps discover why it was suddenly adopted (Carver 1995). The right questions to ask of Mound 1 are therefore not ‘is it Roman?’, ‘is it Swedish?’, or ‘is it Christian?’, but why that, why there, and why then? One day it seems likely that our answers to these questions will become more informative and more subtle than they can be today. They will also reinforce history where this period most needs it, in the affairs of the mind. Sutton Hoo begins with the optimistic expressions of a local warrior class whose political alignment is towards the east: pagan, maritime, and autonomous. By the early seventh century the need for extravagant monumentality becomes pressing and must betray anxieties about threats to intellectual freedom. The complexities, eclecticism, and depth of conjecture revealed by the ship burials are as learned in their way as any Mediterranean codex and possibly more original. That these fears were well grounded is indicated by the subsequent use of the burial ground of kings as a cwealmstow for criminals (Carver 2005: chs. 8, 9, 14).

PORTMAHOMACK: CHOICES OF THE PICTS, SIXTH TO NINTH CENTURIES ......

The third case study is not in Anglo-Saxon England at all but at Portmahomack in the north-east of Scotland (Carver 2008a; Fig. 47.5). It is included to emphasize that the ‘Anglo-Saxons’ were only one group of many expressing ideas in early medieval Britain (see below), and it might be instructive to look at parallel processes and different ideas emerging at the same time as Sutton Hoo but at the other end of the island. At Portmahomack an excavation of 0.75 hectares has unearthed a Pictish settlement which has been designated as monastic, although it is documented by nothing except (somewhat obscurely) an association with St Colman of Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:57 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 945

CarnCarn aa‘¢ BhodaichBhodaich CastlehavenCastlehaven PortPort a ChaisteilChaisteil PortPortPort a’a ChaitCChaithait DORNOCHDORNOCH FFIRTHIRTH Tarbatness WilkhavenWilkhaven ALLANSALLACH BlarBlar á CCathath BINDAL PortmahomackPortmahomack BalnabruachBalnabruach PITFAED

CnocCnoc TighTigh

PITHOGARTYPITHOGARTY PITNELLIESPITNELLIES ARBOLL TARREL

Loch Eye TarrelTarrel GEANIES Portage MOUNTEAGLE PITKERRIE GeaniesGeanies CliffsCliffs CADBOLL Fearn

PortPort aa’ BaintighearnaBBaintighearnaaintighearna HiltonHilton ofof CadbollCadboll Abbot‘sAbbot‘s HavenHaven BalintoreBalintore PortPort anan DruidhDruidh ShandwickShandwick

BayBay ofof NiggNigg Easter Rarichie CROMARTYCROMARTY PITCALNIE FIRTHFIRTH Hill of Nigg

MORAYMORAY FIRTHFIRTH PITCULZEAN 010 1010 kmkm CastleCastle CraigCraig

Possible prehistoric enclosures Beaker burial Current sea level Possible Roman beacon Pictish monuments Sea level in the eighth century Long cist burial(s) Medieval castles Havens Short cist burial(s) Chapels Other burial sites Well PICTISH PLACENAMES Site of battle NORSE PLACENAMES

Figure 47.5 Monastic geography: the Portmahomack excavations (Carver 2008a: ﬁg. 3.17) Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:58 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

946 m. o. h. carver

Lindisfarne, preserved in the Middle Ages and later in its place-name and church dedication. But it has produced a wealth of sculpture paralleled in Iona and the west coast, in southern Pictland, and in Northumbria at least as far south as Yorkshire. One piece carried a Latin inscription in insular majuscules, close to Lindisfarne in type, while another carried Pictish symbols. The settlement began in a small way on top of a hill between the Firth and a marsh in the later sixth century, and its occupants were mainly concerned with obtaining fresh water. At this time the principal investment was in burials, unfurnished but lined with large slabs of stone, a practice that was not specifically Christian but had begun in the local Iron Age. Indeed there were already burials of this kind on the Tarbat peninsula before the first Christian community arrived. In the late seventh or early eighth century the settlement underwent a major development. Large quantities of soil and stones were moved to create a new landscape, with a paved and stone kerbed road running down the hill to a pond formed by building a dam across the valley. On either side of the road were workshops processing cattle and making leather and vellum. Across the valley a building as geometrically perfect as any brooch or gospel book was used by metalworkers engaged in making church vessels of silver, bronze, and glass. The monastic establishment, as indicated by workshops and sculpture that was radiant with Christian symbolism, endured through the eighth century. By the later eighth century, both the Portmahomack churchyard and the Tarbat peninsula as a whole were marked out as sacred spaces with large standing stone crosses. At the end of the eighth century the monastery was burnt down, but revived as a farm and workshops, at first recycling precious metals and latterly (in the ninth to eleventh centuries) as a smithy. The documentary evidence is poor throughout, and particularly for the ninth to eleventh centuries, but a passable attempt can be made at reconciling it with the archaeology. The monastery was probably founded in the late sixth century on a pre-existing Iron Age ‘holy island’ in association with Columba’s visits to the northern Picts. The monastery only exhibited its truly diagnostic parameters in the eighth century when it functioned also as a centre of manufacture, presumably for the equipping of more monasteries. The levels of investment were massive at this time, the standing carved stone monuments erected on the Tarbat peninsula being among the largest and most elaborate known from early medieval Europe. They appear to mark out the monastic territory in the manner known in monastic Ireland (Fig. 47.6; Carver 2008a, b). The destruction of the workshop and its subsequent revival can be placed in the context of the Viking raids and the war that followed between the earls of Orkney and those of Moray, culminating in a major battle in about 1035 at Tarbat Ness itself. However, the intellectual story offered by the archaeology is more interesting than this. In the sixth and early seventh centuries, while the communities of Wasperton and the aristocracy of Sutton Hoo were lavishly burying portable Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:58 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 947

CROSS C

TARBATNESS ROAD TARBAT DISCOVERY CROSS B CROSS A WORKSHOP CENTRE AREA CROSS D

TERRACE WALL

ROAD POND

EARLY ENCLOSURE DITCH WELL

KILN-BARN

ENCLOSURE DITCH 0 m50 m

Figure 47.6 The Tarbat peninsula (Carver 2008a: ﬁg. 9.1)

wealth, the contemporary ideology of Portmahomack was referencing imported Christian ideas moderated by local prehistoric thinking (Carver 2008b), resulting in minimal monumentality But in the eighth century an expansionist phase can be seen in the ﬂowering of monastic industry and the multiplying of cross-slabs on wealthy monastic sites all over northern Britain. The references of the sculpture are Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:58 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

948 m. o. h. carver

to Ireland and Northumbria, of the symbols to Pictland, of the road to the Romans, of the waterworks to the Irish, of the iconography to the Christian Levant, of the inscription to Northumbria. It would be impossible, archaeologically, to assign a primary inﬂuence. The sculpture portrays local ornament, animals, and people with a famous mastery of line, but it also offers metaphysical symbolism and detailed knowledge of the life of King David and the meeting of Anthony and Paul in the desert. On the Tarbat peninsula in the far north-east of Scotland in the eighth century there is no trace of the kind of ignorance or confusion that has earned the dismissive soubriquet ‘syncretism’. These are independent thinkers who are aware of the broad repertoire of contemporary philosophy and chose to give to it their own emphasis. But the thinking and ambition is very different from that of their contemporaries further south.

STAFFORD: INTIMATIONS OF ROMANITAS IN THE TENTH CENTURY ......

The last case study takes to us a burh of the tenth century. Its story in this case is not yet published but the relevant elements are swiftly told. The site of Stafford was a peninsula of cultivated land surrounded by marsh, with a ford across the river Sow. It was the subject of a research campaign from 1975 to 1985 (Fig. 47.7). Excavation at a site in the centre of the town just north of St Mary’s, the principal church in Stafford, showed that grain had been processed there intermittently from the Iron Age through to the twelfth century and beyond. Four tenth-century ovens were found, two for drying and two for baking wheat and oats on a large scale (Moffett 1994). Experiments showed that one of these ovens could produce twenty loaves an hour. To the east was an industrial site, also of the tenth century, producing pottery, the so-called Stafford Ware, with a range of products which included lamps, bowls, and cups (Fig. 47.8). This pottery looks like Trent Vale ware, the local Roman pottery made 700 years earlier in the same place, and of course frequently dug up by Anglo-Saxons; indeed they are so alike, that there is little doubt that the crafts- men were deliberately copying Roman pottery. Severe penalties were imposed on the Stafford potters who got it wrong: amongst the pottery wasters were found three human skulls. The most common form of pot and the only type to be found outside Stafford was the jar, and its numerous examples seem to be all the same size, implying a standard measurement. The context of tenth- and eleventh-century Stafford was the reconquest of Britain by Alfred’s son Edward who took the eastern route, and his sister Æthel- flæda, Lady of the Mercians, who was in command of the left flank. She fortified Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:58 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 949

FRIARS STREET

NEW STREET

STAFFORD: EXCAVATIONS & 12 OBSERVATIONS BROWNING ST

FOREGATE ST Excavation/Recorded strata 7(a) Excavation/Recorded strata 9 (exact location/size unknown) GAOL ROAD Observations N

0 100 200

GAOL STREET QUEENSWAY J

3 D 26 N 20 9, 11 KING‘S POOL MOUNT ST F 42 43 NORTH WALLS CHELL ROAD SALTER ST STAFFORD STREET 23 48 MARKET ST 37

ALBION PL CRABBERY STREET BROAD ST 47 22 29 EASTGATE 18 H BROAD 51 EYE MARTIN ST 50 ST CHAD‘S PL 2 STREET 30 45 33 46 BATH ST I 13 44 5 36 G 34 ST MARY‘s GROVE 32 CLARKE ST b 13 1 GREENGATE STREET 17 14 30 c 25a TIPPING ST a EARL PLACE B 15 27 39 25 16 4 31 d ST MARY‘S PLACE e 10 24 SOUTH WALLS 54 MILL ST GAOL K STREET

TENTERBANKS8 MILL BANK WATER ST 38 55 E

LICHFIELD RD

L A FRIARS ROAD FRIAR‘S WALK

Figure 47.7 Plan of Stafford town, showing the sites of excavation to 1990 (Carver) Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:58:59 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

950 m. o. h. carver

Figure 47.8 Stafford Ware (Carver)

Bridgenorth in 912, Tamworth and Stafford in 913, Hereford in 914, and Runcorn in 915, as well as repairing some old Roman forts. Hereford has provided an excavated example of burh fortiﬁcations, an earth bank eventually revetted in stone. Its rampart lay over a corn-drier dated to the eighth to ninth century, and it is very likely, as H. M. Chadwick suggested 100 years ago, that burhs were sited on places which were already tribute-collecting points; in fact it is most probably the collecting of tribute and storage of food rent that is to be protected by the burh (Chadwick 1905: 255: ‘In earlier times most of the places mentioned in the Burghal Hidage must have been merely royal estates or villages’; cf. Campbell 1986: 109:‘An English villa regia ...was the centre of a fairly wide area all or most of whose people owed something to it’). The distribution of Stafford Ware spreads to the rest of Mercia and its fringes, but only to other burhs (see below; Vince 1985; McCarthy and Brooks 1988: 200). In the marsh at the edge of the Stafford burh, piles of animal bones were found, suggesting the centralized processing of meat. We have here glimpsed a system of the control of resources through a network of burhs, each no doubt with a garrison which was protecting tribute delivered from the estates. The pottery was probably made as a container which formed part of the provisioning strategy, as did the production of loaves. The burh system pioneered by Alfred erected defensive enclosures at twenty-mile intervals in Wessex, and the northward advance of his successors was punctuated by new burhs in a manner that bears more than a passing resemblance to the Roman conquest (Carver 1993: 73). The idea of linked forts was once thought to Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:00 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 951

have been taken, along with many other ideas and works of art, from Carolingian France. But the results from Stafford point rather to an inspiration of the Wessex kings’ own vision of the Roman Empire, and particularly the province of Britannia. Anglo-Saxon kings seem to have noticed the ruins of Roman Britain and the pots that were dug up on old Roman sites when they went there to rob stones to build churches. They became archaeologists themselves. Where we have seen them, the burhs look like Roman forts, both in their shape and distribution. One Wessex burh, Cricklade (1,500 hides, so 2,000 yards around the edge) has been extensively excavated without finding anything inside the defences, suggesting that it never evolved beyond the first stage of a tented encampment (Haslam 1981: 29). Alan Vince, re-examining the archaeological evidence, has shown that the development of the burhs followed a two-phase trajectory (1994). The places fortified in the late ninth and early tenth century have little evidence for manufacture and seem to be military in function. There is a proliferation of mints in the late ninth century, but this can be associated with the conversion of tribute into silver. Winchester is planning its streets in the early tenth century, but even there manufacture does not seem to begin until the mid tenth century. After that we have plenty of evidence for shoe-making, cobbling, pottery-making, glass-working, working precious metals, and iron-smithing. In Wessex at least, the distribution of pottery now suggests a large amount of traffic between the burh and countryside. It is the early eleventh century which sees the flowering of international trade and, incidentally, of deep sea fishing, as suggested by hooks. So, following and elaborating the Vince model, the late ninth- and early tenth- century burh foundations were forts, which protected taxation points and only developed as manufacturing centres in the mid tenth century. They did not participate in international trade, even in London, until the early eleventh century. Stafford is interesting in that it never reached that latter stage of development before the Norman Conquest closed the place down for a century. Archaeologically, that means we can see more clearly the ruthless ‘Roman’ programme of the kings of Wessex, uncluttered by the more open market town that came later. Stafford may be seen as essentially a fort, protecting the delivery of tribute, with a vicus on the east side where pottery was manufactured and cattle were processed.

READING ALIGNMENT ......

The presentation of these four case studies in summary form shows that early medieval sites in Britain can be assigned a number of intellectual properties. The builders of cemeteries, monasteries, and burhs—we could extend this to palaces Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:00 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

952 m. o. h. carver

and wics—could draw on a number of different models of civilization with which they could then declare or deny alignment. Among these we can identify first the prehistoric landscape, an open book from which the Anglo-Saxons could cite, using a commonly understood and inherited vocabulary including chambers, mounds, and ship burials (Williams 1998, 2006). Next, the occupants of south- east Britain could hardly fail to be knowledgeable about the inheritance of the Britons. Although reference here is notoriously elusive, we may find it in the use of certain practices, such as the inclusion of stones in graves, which has a different emphasis countrywide (Carver et al. 2009: chs. 3, 6). Next, the Scandinavian neighbours can be assumed to have had a permanent intellectual presence in eastern Britain, although as with all political alignments there were times when it was embraced and times when it was strenuously rebuffed. It seems likely, although it is less well documented even in Northumbria, that a similar force made itself felt from Ireland. The monastic design, with its water management schemes and the type of territory marked by crosses that declared itself at Portmahomack, may yet prove to underlie Anglo-Saxon monasteries too. The Roman project, of which the Anglo-Saxons were permanently aware, may have acted as the antithesis to the Scandinavian. It can be seen even from this very brief review that archaeology reveals a series of Roman ‘renaissances’ of different kinds in the seventh, eighth, and tenth centuries. In attempting to deconstruct the thinking behind our monuments, we may choose to summarize their elements as Roman, Scandinavian, Christian, or non- Christian. But it will be important to return frequently to the starting point— namely the originality of minds and creative artists at work in the first millennium, who, while they had all these ideas in reach and within their repertoire, need not have surrendered their allegiance to any. Writing of the Iron Age, Barry Cunliffe remarks: ‘The iconography of Britain before the Conquest, reflected largely in the Roman formalisation of the situation, shows that an immense number of local or tribal gods existed, each known by a regional name and each endowed with specific qualities’ (2005: 575–6); at the end of the Roman period, he sees the clock turning back to the Middle Iron Age (2002: 136) so presumably some of this diversity returned. Diversity is certainly implied by the repertoire of pre-Christian burial practice, and even for Christianity James Campbell sees it as a ‘safe generalization’ that ‘England and its church contained much diversity’ (1986: 84; see Brown 2003 for the general phenomenon). Only with the institutionalization of Christian governments and the successful imposition of wide social control would it become possible for these minds to be successfully closed. In Anglo-Saxon England this was probably not achieved or achievable before the eighth century. It will also be noted that if we intend to use archaeology to write an intellectual history of the Anglo-Saxons, it is important that the area of study is not confined to Anglo-Saxon England itself, or even to Britain. Britain is an island that faces onto three seas, each of which had been regularly crossed since the Bronze Age. Each of Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:00 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 953

these seas can be regarded as a thoroughfare, not a barrier (Carver 1990; Cunliffe 2001; 2002: 117). The starting assumption is therefore that, unless prevented from doing so by politics, people and ideas could and did cross them. The references discovered in Anglo-Saxon archaeology require us equally to know the archaeology of British prehistory, the northern Roman Empire, and early medieval Ireland, France, and Scandinavia.

REGIONAL ALLEGIANCE: TERRITORIES, OF THE LAND, OF THE MIND ......

Archaeologists can therefore attempt to deﬁne which kinds of identity and ideology are being expressed by deconstructing references and allusions in the material culture of artefacts, graves, stone monuments, and settlements. It is not expected that they would necessarily agree with each other, any more than an interpretation of Beowulf need command consensus. But we can at least agree to the proposition that the builders of monuments expressed allegiance to a number of different ideas at the same time, even if we cannot agree about the relative emphasis of each. Allegiance to territories is harder to deconstruct since the effect of agency, of expressed intention, is much less certain. Even in modern times, territories cannot easily be created, but emerge out of a claimed ‘natural’ or ‘ethnic’ cohesion. Nevertheless, the expectation should be, as with sites, that the territories represent the resolution of a number of competing claims. They will be strongly inﬂuenced by prehistoric and Roman predecessors, as well as by contemporary government. Documented Anglo-Saxon territories still survive today (East Anglia, Kent, Wessex, Northumbria) so we will not be surprised if these also represent territories of some antiquity. At the same time, by using archaeological evidence, we shall hope to discover otherwise hidden territories, to which people gave their allegiance in some way. We can already anticipate that these different types of territorial allegiance will not always coincide. The historical framework suggests that the early Anglo-Saxon period should map into zones for Angles, Saxons, and Jutes, the middle period into kingdoms and sub-kingdoms as described in the Tribal Hidage, and the late period broadly into the counties and diocese that survived into the twentieth century. However, although no boundaries are known with precision, it can be noted that all these territories map more convincingly onto the Roman and Iron Age territories which preceded them (Fig. 47.9). Kent was the Roman civitas Cantiaci and before that the tribal territory of the Cantii. Mercia, a marchland or border territory, was the border between Britannia Prima and Britannia Secunda, as well as (later) between Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:00 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

954 m. o. h. carver

CORNOVII SMERTAE CAERENI

CARNONACAE VACOMAGI

TAEZALI

DECANTAE

CALEDONES VENICONES

DAMNONII VOTADINI

SELGOVAE NOVANTAE

BRIGANTES BRIGANTES

PARISI PARISI

Occupation ORDOVICES DECEANGLI

CORNOVII CORNOVII CORIELTAUVI ICENI ICENI Military CORIELTAUVI

TRINOVANTES TRINOVANTES DEMETAE DEMETAE CATUVELLAUNI CATUVELLAUNI SILURES SILURES DOBUNNI ATREBATES DOBUNNIATREBATES CANTIACI DUROTRIGES DUROTRIGES CANTIACI

BELGAE REGNENSES DUMNONII BELGAE REGNENSES DUMNONII

PICTS FORTRIU

bbarleyarley

SCOTS ANGLES GODODDIN

kkeyseys BRITONS BERNICIA skinchskinch RHEGED

ANGLES kkingsings

ELMET DEIRA

LINDES FARONA LINDSEY kingskings MYRGENES barleybarley aandnd GWYNEDD LANDES EAST ccrossesrosses MERCIA ANGLIA BRITONS POWYS ANGLES DYFED ccreeree fainitesfainites HWICCE EAST GLYWYSING SEXENA creecree WESSEX CANTWARENA sscribscribs fainitesfainites KENT WEST SEXENA SUSSEX SUTH SEXENA fainitesfainites

SAXONS

BRITONS Figure 47.9 Four territorial mosaics for Britain: (a) Iron Age tribal areas (Cunliffe 2005); (b) Roman Civitates (Wacher 1974: 28); (c) Anglo-Saxon and British kingdoms (Hill 1981: 76, modiﬁed); (d) Truce terms used in twentieth-century school playgrounds (Opie and Opie 1982: 169) Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:01 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 955

England and Wales (Thomas 1981: 200). The territorial divisions of Anglo-Saxon England are basically Iron Age, even if they can sometimes skip several generations. The civitas of the Iceni re-emerges as the diocese of Elmham, although it had been ‘Norfolk’ in between. The mid Saxon kingdom of East Anglia is remembered as Harold’s possession at the Conquest, and Northumbria as Siward’s, although both had been incorporated into the Danelaw in the interim (Hill 1981: 89, 96, 105). Perhaps owed to the geography of language, whether Iron Age or English, these territories and their subdivisions can still come to the surface in unexpected ways, as in Peter and Iona Opie’s famous map of the truce terms used in school playgrounds in the 1950s (Opie and Opie 1982). It excludes the private-school term for fending off attack—pax, a Latin word (like cave) that had survived countrywide but conﬁned to a particular social group. This mapping exercise reinforces the perception that all territories, perhaps in Europe as well as in Britain, are simply the products of the longue dure´e. Gillett (2002: 120) sees terms such as ‘Goths’ as artefacts of late Roman popular geography, nicknames of territories that stuck, ‘with no ethnic connotations whatever’. Sebas- tian Brather feels that the territories that artefacts map onto are merely the previous Roman territories: ‘the differences between Frankish and Alammanian is actually the difference between Roman Gaul and the barbarian lands adjacent to the limes’(2002: 156–7). Not only does the distribution of material culture have no ethnic content, but neither do terms such as Goth, Angle, or Pict at their time of use. They are simply late Roman nicknames given to the current residents of pre- existing long-lived territories. Such territories may have an environmental rationale that political will is powerless to erase. The archaeological investigation not only hopes to endorse these territories but aims to recover hidden ones: popular trends that are working beneath the political radar. The method it uses is primarily the distribution map, which fell out of favour in Britain in the 1960s to such an extent that the production of the excellent Ordnance Survey maps and all its imitations virtually ceased. The reason given was that since all areas of the country had not been investigated to an equal degree of intensity, all distribution maps were ‘misleading’ (Hurst 1976: 288). There has been no such coyness on the Continent and there are at least four good reasons why it would be a shame now to rob ourselves in Britain of this invaluable method of pattern-seeking. First, total representation is impossible for any map and always will be, but that is no reason for not using them, if only to point to areas where further investigation might be desirable. Second, pattern-seeking is what we do; if a pattern fails to survive for more than a few years so be it: nothing is lost except a little vanity. A map is a basis for argument rather than a fact. Third, we have digital mapping technology as never before, so it seems sensible to make use of it. Fourth, we have had twenty years of unparalleled data gathering from CRM mitigation (archaeological investigation in advance of development); not only have we failed to make much use of its distribution to date, but we are presently on the threshold Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:02 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

956 m. o. h. carver

of a data-gathering revolution which will make the previous acquisition look distinctly sparse—namely the advent of the Portable Antiquities Scheme. The distribution mapping that is possible now will reveal and enhance secret systems of allegiance and communication far beyond the tentative exercises given here. Various cultural items have already been mapped and used. Cruciform brooches, square-headed brooches, and pottery have all indicated territories. Old English (but not Latin) inscriptions, sceattas, stycas, and pennies all map territories of use which reflect the relations of Anglo-Saxons and their partners in trade or gift exchange. In sculpture, the ongoing analysis of the Corpus is beginning to discern schools of practice, for example in the concentration of hog backs, or preference for particular kinds of cross, within the general pan-insular dictionary (Cramp 1984). These exercises do not map exactly onto those of the documented territories, so we must at least entertain the possibility that they are following a logic of their own. In the space remaining, I use two examples to explore these themes, both of the late Saxon period. The first example is the late Saxon pottery industry, which John Hurst already warned us not to map (see above); it will nevertheless be worth a try (Fig. 47.10). The territories are to some degree exclusive, so that the supplies relate to a particular zone. John Hurst noticed the ‘strong Roman influence’ in late Saxon pottery and assumed it was due to traditional potters migrating from the Low Countries (1956: 48). But the references can be more local than that. Just as Stafford Ware seems to reflect the same cheerful orange ware of the Trent Valley (or Severn Valley) potters, the Thetford Ware potters echo the dark Roman fabrics of the local Nene Valley. Some of these factors may have their rationale in the local clays. But like the documented territories, the pottery zones may have even deeper roots than the Roman. Some of the Iron Age pottery zones of the sixth to fourth centuries bc published by Barry Cunliffe (Fig. 47.11) prefigure those of the tenth and eleventh centuries ad, for example Darmsden–Linton ¼ Thetford Ware; All Cannings Cross–Meon Hill ¼ Cheddar E.; Chinnor–Wandlebury ¼ St Neots Ware. The St Neots ‘territory’ is prefigured by the early-sixth-century East Anglian contact zone at Wasperton (Fig. 47.2), which may provide a clue to its identity. The distribution actually follows the locus of the Icknield Way, so relating to no political, cultural, or ethnic territory, but simply an ease of communication. The same may be said of Late Saxon Shelly Ware which follows the Thames Valley. Alan Vince (1985: 30) might express surprise that these potters exclusively supplied London, but it is logical if they were bringing their wares downstream by boat. For the late Saxon potters we can therefore discern a zonation that depends, to different degrees, on transport, political investment, and a potting tradition that can lie dormant from the Roman period or even from the Iron Age—up to a thousand years or more. A second example uses an artefact that should not have any prehistoric technical antecedent: the buildings of the Christian church. Although he wrote a large book breaking down Anglo-Saxon architecture into its elements, Harold Taylor was Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:02 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 957

Thetford ware Stamford ware St Neots ware Shelly ware Cheddar E ware Stafford-type ware

0miles 100 0 kilometres 100

Figure 47.10 Late Saxon Pottery distributions, tenth and eleventh centuries (Carv- er after Hurst 1956; 1957; 1958; 1976; Kilmurry 1980; Vince 1985) Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:02 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

958 m. o. h. carver

All Cannings Cross–Meon Hill ditto Somerset variant ditto Dorset variant Park Brow–Caesar‘s Camp Long Wittenham–Allen‘s Pit Chinnor–Wandlebury Darmsden–Linton Highstead–Dollands Moor

0 100 200 kms

Figure 47.11 Iron Age pottery zones, sixth to fourth centuries bc (Cunliffe 2005: 98)

cautious about the significance of their distribution (Taylor 1978: 774). The survival of churches and the survival of the elements themselves were subject to the random forces of later rebuilding. He was thus confident in the regionality of only a few of his architectural attributes, noting the concentration of round towers in Norfolk, the absence of long-short quoins from Northumbria, the stripwork confined to the east coast, apsidal east ends confined (with one exception) south of the Fosse Way, and the clusters of pilaster strip types in Kent and Essex, Norfolk (Fig. 47.12). These examples show that there is some regional variation in tenth- and eleventh-century churches, but do not easily explain it. In some cases, the character of the local stone is a factor (as with the flint pilasters, and possibly the round towers). In others, the ease of communication is probably paramount (as with the east-coast distribution of stone pieces for stripwork). More cultural groupings may emerge with closer analysis, since Taylor’s categories were quite coarse. ‘Stripwork’, for example, includes the gable-headed hoods for windows and doorways that provide such a Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:03 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 959

distinctive aspect of thirteen churches in Norfolk. It is also possible that church- builders, like manuscript-makers, belonged to extensive multi-regional coopera- tives in which ideologies and practices were shared. In any event, if the distribution of some elements (building stone) can have a geological explanation, other preferences (types of opening) probably betray a more intellectual confederation.

CONCLUSION ......

There are therefore at least two kinds of territories for which future early-medieval archaeologists may be disposed to hunt: the one traditional, stable, geographically logical, and probably ancestral, which is reflected in artisan industries; the other innovative, varied, and intellectual, expressed in monumentality. There will be moments when these coincide and moments when they do not; times when only ancestral territories show, and times when these are obliterated by an overwhelming orthodoxy. Up to now, archaeology’s interests have been almost exclusively attracted by the orthodox and imperial, the Christian and the Roman archetypes with their obtrusive and repetitive ‘culture’. Perhaps now is the moment to start listening to the dissident voices. Anglo-Saxon archaeology should not be a private matter for English archaeologists to indulge in. At the scale of the site and of the territory, its new agenda looks to the evidence of prehistory and to the neighbours, both on the island and overseas, to determine the references and allusions being made. We note that the ‘expressivity’, the degree of agency, varies with the kind of material culture. The expression of identity and ideology is strong in high-investment graves and sculpture, lower in pottery and architectural elements. Artefactual territories reflect ancient usage—particularly of routes of communication and prehistoric territories; they seem to say little about documented ethnic groups, and when they do, the documentary labels and the archaeology are often out of synchrony with each other. In the fifth century, when the Angles, Saxons, and Jutes are supposed to have partitioned the country, there is little demarcation of the graves. It becomes marked only in the following generations, during the sixth century (Hills 2003: 106). In the seventh century, when the Anglo-Saxon zone is theoretically divided into kingdoms, the grave goods not only take on a distinctly Roman or Byzantine quality, but proclaim a united territory which equates with later England (Geake 1997: 1999; cf. Campbell 1986: 67). But when we arrive in the English kingdom of the tenth century, the increase in prosperity and in material engagement lights up the local taste, and reveals artefactual zones that are variations on those of the mid Saxon period or its Roman and prehistoric predecessors. Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:03 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

960 m. o. h. carver

Stone pilaster strips Long and short quoins Stripwork around belfry windows

0miles 100 0 kilometres 100

Figure 47.12 Architectural preferences in Anglo-Saxon churches, tenth and eleventh centuries (Carver after Taylor 1978: 890, 920, 930, 945) Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:04 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 961

For the thinking expressed in sites and monuments, territory, real and perceived, is one of the wells on which creative monumentality can draw. Stafford was aware of its Iron Age function as a processor of tribute, but acknowledged the previous Roman territory and the supply zone of the Roman potters as well as the contemporary territory of Æthelﬂæd’s conquest. This is a highly intellectual agenda of some complexity, but it is not impenetrable or even forbidding. The Anglo-Saxons were an intellectual and complex people. They deserve more from their archaeologists than superﬁcial explanations based on race, social structure, and Chris- tianity. Our business is to release and celebrate the diversity of the times, its promises and menaces, as the earliest English themselves felt it to be.

ACKNOWLEDGEMENTS ......

The basis for selecting the case studies discussed here is only that the author knows them well. There are surely better ones and will be still more in the future. I am grateful to Nicky Toop of FAS Ltd for making the maps.

REFERENCES

Andre´n, A. (1998). Between Artifacts and Texts: Historical Archaeology in Global Perspective. New York: Plenum Press. Bartholomew, P. (2005). ‘Continental connections: Angles, Saxons and others in Bede and Procopius’. Anglo-Saxon Studies in Archaeology and History 13: 19–30. Bassett, S. (ed.) (1989). The Origins of Anglo-Saxon Kingdoms. London and New York: Leicester University Press. Blair, J. (2005). The Church in Anglo-Saxon Society. Oxford: Oxford University Press. Bradley, R. (1987). ‘Time regained: the creation of continuity’. Journ. Brit. Archaeol. Assoc. 140: 1–17. Brather, S. (2002). ‘Ethnic identities as constructions of archaeology: the case of the Alamanni’, in A. Gillett (ed.), On Barbarian Identity: Critical Approaches to Ethnicity in the Early Middle Ages. Turnhout: Brepols, 149–75. Braudel, F. (1972). The Mediterranean and the Mediterranean World in the Age of Philip II, Volume 1. London: Collins. Brown, P. (2003). The Rise of Western Christendom: Triumph and Diversity AD 200–1000. 2nd edition. Oxford: Blackwell. Bruce-Mitford,R.L.S.(1975). The Sutton Hoo Ship Burial, Volume 1. London: British Museum. Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:04 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

962 m. o. h. carver

Brugmann, B., and Penn, K., with K. Hilund Nielsen (2007). Aspects of Anglo-Saxon Burial: Morning Thorpe, Spong Hill, Bergh Apton and Westgarth Gardens. East Anglian Archaeology 119. Gressenhall: Norfolk Museums Service. Buck, C. E., Cavanagh, W. G., and Litton, C. D. (1996). Bayesian Approach to Interpreting Archaeological Data. Chichester: Wiley. Budd, P., Millard, A., Chenery, C., Lucy, S., and Roberts, C. (2004). ‘Investigating populations movement by stable isotope analysis: a report from Britain’. Antiquity 78: 127–41. Campbell, J. (1986). Essays in Anglo-Saxon History. London and Ronceverte: Hambledon Press. Carver,M.O.H.(1989). ‘Kingship and material culture in early Anglo-Saxon East Anglia’, in Bassett (ed.), Origins, 141–58. —— (1990). ‘Pre-Viking trafﬁc in the north sea’, in S. McGrail (ed.), Maritime Celts, Frisians and Saxons. CBA Research Report 71. York: Council for British Archaeology, 117–25. —— (ed.) (1992). The Age of Sutton Hoo. Woodbridge: Boydell Press. —— (1993). Arguments in Stone: Archaeology and the European Town in the First Millenni- um. Oxford: Oxbow Books. —— (1995). ‘Ship burial in early Britain: ancient custom or political signal?’, in O. Crumlin-Pedersen and B. Munch Thye (eds.), The Ship as Symbol in Prehistoric and Medieval Scandinavia. Copenhagen: The National Museum, 111–24. —— (1998). Sutton Hoo: Burial Ground of Kings. London: British Museum. —— (2000). ‘Burial as poetry: the context of treasure in Anglo-Saxon graves’, in E. Tyler (ed.), Treasure in the Medieval West. Woodbridge: Boydell, 25–48. —— (2001). ‘Why that, why there, why then? The politics of early medieval monumentality’, in A. Macgregor and H. Hamerow (eds.), Image and Power in Early Medieval British Archaeology. Essays in Honour of Rosemary Cramp. Oxford: Oxbow Books, 1–22. —— (2002a). ‘Reﬂections on the meanings of monumental barrows in Anglo-Saxon England’, in S. Lucy and A. Reynolds (eds.), Burial in Early Medieval England and Wales. Society for Medieval Archaeology Monograph Series 17. London: Society for Medieval Archaeology, 132–43. —— (2002b). ‘Marriages of true minds: archaeology with texts’, in B. Cunliffe, W. Davies, and C. Renfrew (eds.), Archaeology: The Widening Debate. London: The British Academy, 465–96. —— (ed.) (2003). The Cross goes North: Processes of Conversion in Northern Europe, AD 300–1300. York: York Medieval Press. —— (2005). Sutton Hoo: A Seventh-Century Princely Burial Ground and its Context. Reports of the Research Committee of the Society of Antiquaries of London 69. London: The British Museum Press. —— (2008a). Portmahomack: Monastery of the Picts. Edinburgh: Edinburgh University Press. —— (2008b). The Pictish Monastery at Portmahomack. Jarrow Lecture. —— Hills, C., and Scheschkewitz, J. (2009). Wasperton: A Roman, British and Anglo- Saxon Cemetery in Central England. Woodbridge: Boydell Press. Chadwick, H. M. (1905). Studies on Anglo-Saxon Institutions. Cambridge: Cambridge University Press. Collingwood, R. G., and Myres,J.N.L.(1936). Roman Britain and the English Settle- ments. Oxford: Oxford University Press. Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:05 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 963

Cramp, R. J. (1984). A Corpus of Anglo-Saxon Stone Sculpture. Oxford: Oxford University Press. Cunliffe, B. (2001). Facing the Ocean: The Atlantic and its Peoples. Oxford: Oxford University Press. —— (2002). ‘Tribes and empires c.1500 BC–AD 500’, in P. Slack and R. Ward (eds.), The Peopling of Britain: The Shaping of a Human Landscape. The Linacre Lectures 1999. Oxford: Oxford University Press, 115–38. —— (2005). Iron Age Communities in Britain. 4th edition. London: Routledge. Filmer-Sankey, W., and Pestell, T. (2001). Snape Anglo-Saxon Cemetery: Excavations and Surveys 1824–1992. East Anglian Archaeology 95. Gressenhall: Norfolk Museums Service. Fletcher, R. (1997). The Conversion of Europe: From Paganism to Christianity 371–1386 AD. London: Harper Collins. Frank, R. (1992). ‘Beowulf and Sutton Hoo: The odd couple’, in C. B. Kendall and P. S. Wells (eds.), Voyage to the Other World: The Legacy of Sutton Hoo. Minneapolis: Univer- sity of Minnesota Press, 47–64. Frazer, W. O., and Tyrrell, A. (eds.) (2000). Social Identity in Early Medieval Britain. Leicester: Leicester University Press. Geake, H. (1997). The Use of Grave-Goods in Conversion-Period England, c. 600–c. 850. BAR British Ser. 261. Oxford: British Archaeological Reports. —— (1999) ‘Invisible kingdoms: the use of grave-goods in seventh-century England’. Anglo-Saxon Studies in Archaeology and History 10: 203–15. Gillett, A. (ed.) (2002). ‘Was ethnicity politicised in the earliest medieval kingdoms?’, in A. Gillett (ed.), On Barbarian Identity: Critical Approaches to Ethnicity in the Early Middle Ages. Turnhout: Brepols, 85–121. Ha¨rke, H. (2002). ‘Kings and warriors. Population and landscape from post-Roman to Norman Britain’, in P. Slack and R. Ward (eds.), The Peopling of Britain: The Shaping of a Human Landscape. The Linacre Lectures 1999. Oxford: Oxford University Press, 145–75. —— (2007) ‘Ethnicity, “race” and migration in mortuary archaeology: an attempt at a short answer’. Anglo-Saxon Studies in Archaeology and History 14: 12–18. Haslam, J. (1981). ‘Cricklade’, in J. Schoﬁeld and D. Palliser (eds.), Recent Archaeological Research in English Towns. York: Council for British Archaeology, 28–30. Higham, Nick (1995). An English Empire: Bede and the early Anglo-Saxon Kings. Manche- ster: Manchester University Press. Hill, D. (1981). An Atlas of Anglo-Saxon England. Oxford: Blackwell. Hills, C. (2003). Origins of the English. London: Duckworth. —— (2007). ‘History and archaeology: the state of play in early medieval Europe’. Antiquity 81: 191–200. Hines, J., Hilund Nielsen, K., and Siegmund, F. (eds.) (1999). The Pace of Change: Studies in Early-Medieval Chronology. Oxford: Oxbow Books. Hodges, R. (1989). The Anglo-Saxon Achievement. London: Duckworth. Hurst, J. G. (1956). ‘Saxo-Norman pottery in East Anglia’. Proceedings of the Cambridge Antiquarian Society 49: 43–70. —— (1957). ‘Saxo-Norman pottery in East Anglia. Part II. Thetford Ware with an account of Middle Saxon Ipswich ware (with S E West)’. Proceedings of the Cambridge Antiquarian Society 50: 29–60. —— (1958). ‘Saxo-Norman pottery in East Anglia Part III. Stamford Ware’. Proceedings of the Cambridge Antiquarian Society 51: 37–65. Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:05 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

964 m. o. h. carver

—— (1976). ‘The pottery’, in D. M. Wilson (ed.), The Archaeology of Anglo-Saxon England. London: Methuen, 283–348. Kilmurry, K. (1980). The Pottery Industry of Stamford, Lincolnshire c AD 850–1250: Its Manufacture, Trade and Relationship with Continental Wares, with a Classiﬁcation and Chronology. BAR British Series 84. Oxford: British Archaeological Reports. Levison, W. (1946). England and the Continent in the Eighth Century. Oxford: Clarendon Press. Lucy, S. (2000). The Anglo-Saxon Way of Death: Burial Rites in Early England. Stroud: Sutton Publishing. Mccarthy, M. R., and Brooks, C. M. (1988). Medieval Pottery in Britain AD 900–1600. Leicester: Leicester University Press. Moffett, L. (1994). ‘Charred cereals from some ovens/kilns in late Saxon Stafford and the botanical evidence for the pre-burh economy’, in J. Rackham (ed.), Environment and Economy in Anglo-Saxon England. CBA Research Report 89. York: Council for British Archaeology, 55–64. Moreland, J. (2000). ‘Ethnicity, power and the English’, in Frazer and Tyrrell (eds.), Social Identity in Early Medieval Britain, 23–51. Mu¨ller-Wille,M.(1970–71). ‘Pferdegrab und Pferdeopfer im fru¨hen Mittelalter’. Berich- ten v.d. Rijksdienst v.h. Oudheidkundig Bodemonderzoek 20–21: 119–248. —— (1974). ‘Boat-graves in Northern Europe’. International Journal of Nautical Archaeolo- gy 3: 187–204. —— (1992). ‘Monumentale Grabhu¨gel der Vo¨lkerwanderungszeit in Mittel- und Nordeur- opa. Bestand und Deutung’,in W. Paravicini (ed.), Mare Balticum: Beitra¨ge zur Geschichte des Ostseeraums im Mittelalter und Neuzeit. Festschift zum 65. Geburtstag von Erich Hoffman. Kieler Historische Studien, Band 36, 1–20. —— (1993). ‘Burial mounds and burial practices’, in P. Pulsiano (ed.), Medieval Scandina- via. New York and London: Garland, 58–60. —— (1995). ‘Boat-graves, old and new views’, in O. Crumlin-Pedersen (ed.), The Ship as Symbol in Prehistoric and Medieval Scandinavia. Studies in Archaeology and History 1. Copenhagen: The National Museum, 101–10. —— (1998). ‘Zwei religio¨se Welten: Bestattungen der fra¨nkischen Ko¨nige Childerich und Chlodwig’. Abhandlungen der Akademie der Wissenschaften und Literatur. Mainz–Stutt- gart: Steiner: Geistes- und sozialwissenschaftliche Klasse 1: 3–45. Opie, I. and Opie, P. (1982). The Lore and Language of Schoolchildren. Granada. Pryor, F. (2004). Britain AD. London: Harper Collins. Rahtz, P., Dickinson, T. M., and Watts, L. (eds.) (1980). Anglo-Saxon Cemeteries 1979. BAR British Series 82. Oxford: British Archaeological Reports. —— Hirst, S., and Wright, S. M. (2000). Cannington Cemetery. Britannia Monograph Series 17. London: Society for the Promotion of Roman Studies. Renfrew, C., and Cherry, J. F. (eds.) (1986). Peer Polity Interaction and Socio-Political Change. Cambridge: Cambridge University Press. Taylor, H. M. (1978). Anglo-Saxon Architecture, Volume 3. Cambridge: Cambridge Univer- sity Press. Thomas, C. (1981). Christianity in Roman Britain to AD 500. London: Batsford. Thomas, M. G., Stumpf, M. P. H., and Ha¨rke, H. (2006). ‘Evidence for an apartheid-like social structure in early Anglo-Saxon England’. Proceedings of the Royal Society B. doi:10.1098/rspb.2006.3627: 1–7. Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:05 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi

intellectual territories in anglo-saxon england 965

Vince, A. G. (1985) ‘The Saxon and medieval pottery of London: a review’. Medieval Archaeology 29: 25–93. —— (1994). ‘Saxon urban economies: an archaeological perspective’, in J. Rackham (ed.), Environment and Economy in Anglo-Saxon England. CBA Research Report 89. York: Council for British Archaeology, 108–19. Wacher, J. (1974). The Towns of Roman Britain. London: Batsford. Welch, M. (1992). Anglo-Saxon England. London: English Heritage. Whitelock, D. (1954). The Beginnings of English Society. Harmondsworth: Penguin. Williams, H. (1998). ‘Ancient landscape and the dead: the reuse of prehistoric and Roman monuments as early Anglo-Saxon burial sites’. Medieval Archaeology 41: 1–32. —— (2005). ‘Heathen graves and Victorian Anglo-Saxonism: Assessing the Archaeology of John Mitchell Kemble’. Anglo-Saxon Studies in Archaeology and History 13: 1–18. —— (2006). Death and Memory in Early Medieval Britain. Cambridge: Cambridge University Press. Woolf, A. (2000) ‘Community, identity and kingship in Early England’, in Frazer and Tyrrell (eds.), Social Identity in Early Medieval Britain, 91–110. Comp. by: PG2448 Stage : Proof ChapterID: 0001201373 Date:8/9/10 Time:20:59:05 Filepath:d:/womat-filecopy/0001201373.3D OUP UNCORRECTED PROOF – REVISES, 8/9/2010, SPi